| 摘要 |
Method for recovering a relative depth map from a single image or a sequence of still images, comprising:
- decomposing the image into sets U » , each formed by pixels whose intensity is in a predefined interval;
- decomposing sets U » into connected components C i » , a subset of adjacent pixels according to a connectivity rule;
- for each connected component C i » , computing membership measures indicative of the probability of each pixel x ˆ‰ C i » to belong to it and being a function of:
€¢ the distance C i » ) from the pixel x to said connected component C i » ;
€¢ the curvature º x �¢ C i » of the level line of the distance function C i » ) from the pixel x to said connected component C i » ;
- for each pixel x of the image, compute a local depth feature Z(x) indicative of the probability of the pixel x belonging simultaneously to different connected components and which provides a relative depth order estimation at pixel x . |
